Gauging consumer preferences regarding algorithm-controlled products

Global demand for smart products, driven by algorithms and AI is growing. A study recently published in the Journal of Consumer Research investigates how people respond to AI-driven products like voice assistants, robotic vacuum cleaners and smart refrigerators. Specifically, the research compares two kinds of products: those with high-adaptivity algorithms that can learn and adjust, and those with low-adaptivity algorithms that are pre-programmed with no learning ability. The key question is: Do consumers prefer one type over the other, and does this translate into different product choices?

Adaptive and pre-programmable products

In their study entitled Unveiling the Mind of the Machine, researcher Emanuel de Bellis (University of St.Gallen), together with his co-authors Melanie Clegg (WU Vienna), Reto Hofstetter (University of Lucerne), and Bernd Schmitt (Columbia Business School), note that the key findings indicate in general that consumers prefer products equipped with high-adaptivity algorithms. However, this preference is not universal and hinges on the desired range of outcomes a product is expected to deliver. 

Marketing intelligent products intelligently

For example, products with many possible outcomes (like a voice assistant), consumers prefer high-adaptivity algorithms because these algorithms are seen as more creative. By contrast, consumers favor low-adaptivity algorithms for products with a limited number of outcomes (like a smart lock) because these algorithms are seen as more predictable. 

In essence, this research sheds light on the significant role that different types of algorithms play in how consumers perceive and prefer products. It underlines the consequences of revealing the “mind” of the machine to consumers, providing valuable insights for the marketing of algorithm-controlled consumer goods.

Image: Adobe Stock / FutureFocus